A 22-point scorer in Big V is not a 22-point scorer in NBL1. Everyone knows that. Almost nobody adjusts for it properly, because the adjustment people reach for is a discount, and a discount is the wrong tool.
I keep a multi-league database because I got tired of comparing players across competitions in my head. NBL1 men and women, Big V, WNBL, in one place, queryable. The first thing it taught me was that the question I had been asking was wrong. "How many points will she score for us" has no answer. "What does she do, how often, and does that survive a harder league" has three, and they are the ones that matter.
Volume does not transfer. Rate does. Points, rebounds and assists per game are a function of minutes and touches, and both of those are properties of the team she left, not of her. Move a Big V primary into an NBL1 rotation and her usage falls off a cliff before she has done anything right or wrong. Per-minute and per-possession rates survive the move far better because they describe her, not her old role. Start there.
Percentiles are league-local and people forget it constantly. An 88th percentile finisher in Big V and an 88th percentile finisher in NBL1 are not the same player, because the two distributions do not overlap. The percentile tells you where she sits among the people she played against. It says nothing about the people she would be playing against next season. Any cross-league percentile comparison needs the leagues named next to it or it is worse than no number at all.
Efficiency at volume is the thing that actually breaks. A player can hold her shooting percentage moving up a level if her shot profile gets easier at the same time, which is exactly what happens when she stops being the first option. That looks like she translated. She did not. Her efficiency held because her difficulty dropped. The honest read is to look at what she shot on contested attempts and off the dribble, separately, and ask whether the harder league will let her keep the easy ones.
Role is the strongest predictor and it is not in the box score. The recruit who was her old team's only creator has been solving a problem for two years that our team may not need solved. The recruit who was the fourth option on a good side has been doing something closer to what we are asking. Neither is better. They are different, and the mismatch is where most imports go wrong: we sign the numbers of the first player and the job of the second.
What I actually look at first
Shape before size. Roughly in this order:
Where do her shots come from. Rim, mid, corner three, above the break. That distribution is a personality trait and it moves less than anything else on the sheet.
What she does when the first option dies. Every league has open threes. Fewer have players who make the second decision. This is the single most transferable skill I know of and the box score does not carry it, so it is a film question.
Defensive position and defensive load. Who did she guard, and was she hidden. A player hidden on the weakest offensive threat for a season has an unread defensive profile, not a good one.
Availability. Games played, and the pattern of the games missed. This is where I break my own rule and use a counting stat, because it is the only one where the raw number is the point.
What the data will not tell you, and I will not pretend otherwise
Age, contract status and availability for next season are not in any of these datasets in a form I would put in front of a coach. Neither is why she left, what her body has been through, or whether she wants to be here. My database will rank her against 900 other players in four leagues. It will not tell you whether she picks up the phone.
That is not a gap I am trying to close with more data. It is the part where somebody who knows the league picks up the phone. The database is there to make sure that call goes to the right eight players instead of the wrong forty.
The one thing I would change about how most clubs do this
Stop shortlisting on a stat line and then watching film to confirm it. Do it the other way. Shortlist on shape, then use the numbers to check whether the shape held up over a season rather than over the four clips somebody sent you. Film first is how you fall in love with a highlight. Numbers first is how you sign an average. Shape first, checked both ways, is how you sign a player who fits.